You are on page 1of 8

# Discovering Statistics Using SPSS: Chapter 15

## Chapter 15: Answers

The University of Sussex is constantly seeking to employ the best people possible as lecturers
(no, really, it is). Anyway, they wanted to revise a questionnaire based on Bland’s theory of
research methods lecturers. This theory predicts that good research methods lecturers should
have four characteristics: (1) a profound love of statistics; (2) an enthusiasm for experimental
design; (3) a love of teaching; and (4) a complete absence of normal interpersonal skills.
These characteristics should be related (i.e. correlated). The ‘Teaching Of Statistics for
Scientific Experiments’ (TOSSE) already existed, but the university revised this questionnaire
and it became the ‘Teaching Of Statistics for Scientific Experiments — Revised’ (TOSSE—R).
The gave this questionnaire to 239 research methods lecturers around the world to see if it
supported Bland’s theory. The questionnaire is below and the data are in TOSSE-R.sav.
Conduct a factor analysis (with appropriate rotation) to see the factor structure of the data.

SD D N A SA

## I once woke up in the middle of a vegetable patch hugging a turnip

1 A A A A A
that I'd mistakenly dug up thinking it was Roy's largest root

2
If I had a big gun I'd shoot all the students I have to teach A A A A A

## 3 I memorize probability values for the F-distribution A A A A A

4
I worship at the shrine of Pearson A A A A A

## Teaching others makes me want to swallow a large bottle of bleach

6 because the pain of my burning oesophagus would be light relief in A A A A A
comparison

9 A A A A A
morning

11 A A A A A
rotation

## Dr. Andy Field Page 1 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

13
Designing experiments is fun A A A A A

14 A A A A A
the pub

15 A A A A A
Analysis

16 A A A A A
thrills me

## I enjoy sitting in the park contemplating whether to use participant

17 A A A A A
observation in my next experiment

18 A A A A A
my bowels

20 A A A A A
individual

A A A A A
my groin

## I quiver with excitement when thinking about designing my next

22 A A A A A
experiment

I often spend my spare time talking to the pigeons ... and even
23 A A A A A
they die of boredom

## I tried to build myself a time machine so that I could go back to the

24 1930s and follow Fisher around on my hands and knees licking the A A A A A
floor on which he'd just trodden

25 I love teaching A A A A A

## I love teaching because students have to pretend to like me or

27 A A A A A
they'll get bad marks

## 28 My cat is my only friend A A A A A

SPSS Output
Correlation Matrixa

a. Determinant = 1.498E-06

Multicollinearity: The determinant of the correlation matrix was 0.00000149, which is smaller
than 0.00001 and, therefore, indicates that multicollinearity could be a problem in these data
(although, strictly speaking, because we’re using principal components analysis we don’t need
to worry).

## Dr. Andy Field Page 2 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

## KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling

## Bartlett's Test of Approx. Chi-Square 3015.282

Sphericity df 378
Sig. .000

Communalities

Initial Extraction
I once woke up in the middle of a vegetable patch hugging a turnip that I'd mistakenly dug up thinking it was
1.000 .631
Roy's largest root
If I had a big gun I'd shoot all the students I have to teach 1.000 .641
I memorize probability values for the F-distribution 1.000 .558
I worship at the shrine of Pearson 1.000 .568
I still live with my mother and have little personal hygiene 1.000 .561
Teaching others makes me want to swallow a large bottle of bleach because the pain of my burning
1.000 .638
oesophagus would be light relief in comparison
Helping others to understand Sums of Squares is a great feeling 1.000 .484
I like control conditions 1.000 .682
I calculate 3 ANOVAs in my head before getting out of bed every morning 1.000 .632
I could spend all day explaining statistics to people 1.000 .409
I like it when people tell me I've helped them to understand factor rotation 1.000 .542
People fall asleep as soon as I open my mouth to speak 1.000 .263
Designing experiments is fun 1.000 .521
I'd rather think about appropriate dependent variables than go to the pub 1.000 .696
I soil my pants with excitement at the mere mention of Factor Analysis 1.000 .498
Thinking about whether to use repeated or independent measures thrills me 1.000 .676
I enjoy sitting in the park contemplating whether to use participant observation in my next experiment 1.000 .709
Standing in front of 300 people in no way makes me lose control of my bowels 1.000 .519
I like to help students 1.000 .540
Passing on knowledge is the greatest gift you can bestow an individual 1.000 .480
Thinking about Bonferroni corrections gives me a tingly feeling in my groin 1.000 .557
I quiver with excitement when thinking about designing my next experiment 1.000 .763
I often spend my spare time talking to the pigeons ... and even they die of boredom 1.000 .605
I tried to build myself a time machine so that I could go back to the 1930s and follow Fisher around on my
1.000 .653
hands and knees licking the floor on which he'd just trodden
I love teaching 1.000 .541
I spend lots of time helping students 1.000 .593
I love teaching because students have to pretend to like me or they'll get bad marks 1.000 .625
My cat is my only friend 1.000 .561
Extraction Method: Principal Component Analysis.

Sample Size: McCallum et al (1999) have demonstrated that when communalities after
extraction are above .5, a sample size between 100-200 can be adequate and even when
communalities are below .5 a sample size of 500 should be sufficient. We have a sample size
of 239 with some communalities below .5, and so the sample size may not be adequate.
However, the KMO measure of sampling adequacy is .894, which is above Kaiser’s (1974)
recommendation of .5. This value is also ‘meritorious’ (and almost ‘marvelous’) according to
Hutcheson & Sofroniou (1999). As such, the evidence suggests that the sample size is
adequate to yield distinct and reliable factors.
Bartlett’s test: This tests whether the correlations between questions are sufficiently large for
factor analysis to be appropriate (it actually tests whether the correlation matrix is sufficiently
different from an identity matrix). In this case it is significant (χ2 (378) = 2989.77, p < .001)
indicating that the correlations within the R-matrix are sufficiently different from zero to
warrant factor analysis.

## Dr. Andy Field Page 3 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

## Initial Eigenvalues Extraction Sums of Squared Loadings Rotation

Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total
1 9.028 32.243 32.243 9.028 32.243 32.243 6.923
2 2.750 9.823 42.066 2.750 9.823 42.066 3.752
3 1.709 6.102 48.169 1.709 6.102 48.169 3.693
4 1.504 5.370 53.539 1.504 5.370 53.539 5.238
5 1.155 4.127 57.665 1.155 4.127 57.665 3.159
6 .975 3.482 61.147
7 .921 3.289 64.436
8 .828 2.957 67.394
9 .789 2.819 70.212
10 .734 2.620 72.832
11 .708 2.529 75.362
12 .658 2.350 77.711
13 .627 2.240 79.951
14 .584 2.085 82.036
15 .549 1.960 83.996
16 .539 1.925 85.921
17 .492 1.757 87.678
18 .453 1.619 89.297
19 .412 1.471 90.768
20 .399 1.426 92.195
21 .341 1.218 93.412
22 .336 1.200 94.612
23 .314 1.122 95.734
24 .289 1.033 96.767
25 .271 .966 97.733
26 .255 .911 98.644
27 .212 .759 99.403
28 .167 .597 100.000
Extraction Method: Principal Component Analysis.
a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

Scree Plot
10

4
Eigenvalue

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27

Component Number

## Dr. Andy Field Page 4 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

Extraction: SPSS has extracted 5 factors based on Kaiser’s criterion of retaining factors with
eigenvalues greater than 1. Is this warranted? Kaiser’s criterion is accurate when there are
less than 30 variables and the communalities after extraction are greater than .7, or when the
sample size exceeds 250 and the average communality is greater than .6. For these data the
sample size is 239, there are 28 variables, and the mean communality is .577 so extracting 5
factors is not really warranted. The scree plot shows clear inflexions at 3 and 5 factors and so
using the scree plot you could justify extracting 3 or 5 factors.

Pattern Matrixa

Component
1 2 3 4 5
Thinking about whether to use repeated or independent measures thrills me .829
I'd rather think about appropriate dependent variables than go to the pub .813
I quiver with excitement when thinking about designing my next experiment .765
I enjoy sitting in the park contemplating whether to use participant observation in
.733
my next experiment
Designing experiments is fun .575
I like control conditions .556
I could spend all day explaining statistics to people .458
I calculate 3 ANOVAs in my head before getting out of bed every morning
I like to help students .735
Passing on knowledge is the greatest gift you can bestow an individual .651
I love teaching .633
I love teaching because students have to pretend to like me or they'll get bad
.556
marks
Helping others to understand Sums of Squares is a great feeling .470
I spend lots of time helping students .461
I like it when people tell me I've helped them to understand factor rotation
I often spend my spare time talking to the pigeons ... and even they die of
.713
boredom
My cat is my only friend .702
I still live with my mother and have little personal hygiene .692
People fall asleep as soon as I open my mouth to speak .493
I tried to build myself a time machine so that I could go back to the 1930s and
follow Fisher around on my hands and knees licking the floor on which he'd just .752
trodden
I memorize probability values for the F-distribution .662
I worship at the shrine of Pearson .589
I soil my pants with excitement at the mere mention of Factor Analysis .481
Thinking about Bonferroni corrections gives me a tingly feeling in my groin .433 .479
I once woke up in the middle of a vegetable patch hugging a turnip that I'd
.445
mistakenly dug up thinking it was Roy's largest root
Teaching others makes me want to swallow a large bottle of bleach because
.818
the pain of my burning oesophagus would be light relief in comparison
If I had a big gun I'd shoot all the students I have to teach .784
Standing in front of 300 people in no way makes me lose control of my bowels .549
Extraction Method: Principal Component Analysis.
Rotation Method: Oblimin with Kaiser Normalization.
a. Rotation converged in 15 iterations.

Rotation: You should chose an oblique rotation because the question says that the constructs
we’re measuring are related.
Looking at the pattern matrix (and using loadings greater than .4 as recommended by
Stevens) we see the following pattern:
Factor 1:
• Q 16. Thinking about whether to use repeated or independent measures thrills me
• Q 14. I'd rather think about appropriate dependent variables than go to the pub
• Q 22. I quiver with excitement when thinking about designing my next experiment

## Dr. Andy Field Page 5 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

• Q 17. I enjoy sitting in the park contemplating whether to use participant observation in
my next experiment
• Q 13. Designing experiments is fun
• Q 8. I like control conditions
• Q 10. I could spend all day explaining statistics to people
Factor 2:
• Q 9. I calculate 3 ANOVAs in my head before getting out of bed every morning
• Q 19. I like to help students
• Q 20. Passing on knowledge is the greatest gift you can bestow an individual
• Q 25. I love teaching
• Q 27. I love teaching because students have to pretend to like me or they'll get bad
marks
• Q 7. Helping others to understand Sums of Squares is a great feeling
• Q 26. I spend lots of time helping students
Factor 3:
• Q 23. I often spend my spare time talking to the pigeons ... and even they die of
boredom
• Q 28. My cat is my only friend
• Q 5. I still live with my mother and have little personal hygiene
• Q 12. People fall asleep as soon as I open my mouth to speak
Factor 4:
• Q 24. I tried to build myself a time machine so that I could go back to the 1930s and
follow Fisher around on my hands and knees licking the floor on which he'd just trodden
• Q 3. I memorize probability values for the F-distribution
• Q 4. I worship at the shrine of Pearson
• Q 15. I soil my pants with excitement at the mere mention of Factor Analysis
• Q 21. Thinking about Bonferroni corrections gives me a tingly feeling in my groin
• Q 1. I once woke up in the middle of a vegetable patch hugging a turnip that I'd
mistakenly dug up thinking it was Roy's largest root
Factor 5:
• Q 6. Teaching others makes me want to swallow a large bottle of bleach because the
pain of my burning oesophagus would be light relief in comparison
• Q 2. If I had a big gun I'd shoot all the students I have to teach
• Q 18. Standing in front of 300 people in no way makes me lose control of my bowels

No Factor:
• Q 11. I like it when people tell me I've helped them to understand factor rotation

Factor 1 seems to relate to research methods, factor 2 to teaching, factor 3 to general social
skills, factor 4 to statistics and factor 5 to, well, err, teaching again. All in all this isn’t

## Dr. Andy Field Page 6 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

particularly satisfying and doesn’t really support the four factor model. We saw earlier that the
extraction of 5 factors probably wasn’t justified. In fact the scree plot seems to indicate 3.
Let’s re-run the analysis but asking SPSS for three factors. Let’s see how this changes the
pattern matrix:
Pattern Matrixa

Component
1 2 3
I quiver with excitement when thinking about designing my next experiment .799
I like control conditions .794
I enjoy sitting in the park contemplating whether to use participant observation in my next experiment .788
Thinking about Bonferroni corrections gives me a tingly feeling in my groin .745
Designing experiments is fun .715
I calculate 3 ANOVAs in my head before getting out of bed every morning .682
I memorize probability values for the F-distribution .669
I once woke up in the middle of a vegetable patch hugging a turnip that I'd mistakenly dug up thinking it was
.623
Roy's largest root
I tried to build myself a time machine so that I could go back to the 1930s and follow Fisher around on my hands
.620
and knees licking the floor on which he'd just trodden
I worship at the shrine of Pearson .615
Thinking about whether to use repeated or independent measures thrills me .583
Helping others to understand Sums of Squares is a great feeling .568
I soil my pants with excitement at the mere mention of Factor Analysis .542
I like it when people tell me I've helped them to understand factor rotation .535
I could spend all day explaining statistics to people .443
I'd rather think about appropriate dependent variables than go to the pub .437
I like to help students .618
If I had a big gun I'd shoot all the students I have to teach -.560
Teaching others makes me want to swallow a large bottle of bleach because the pain of my burning
-.531
oesophagus would be light relief in comparison
Standing in front of 300 people in no way makes me lose control of my bowels -.519
I spend lots of time helping students .406 .515
I love teaching .449
Passing on knowledge is the greatest gift you can bestow an individual
I still live with my mother and have little personal hygiene .719
I often spend my spare time talking to the pigeons ... and even they die of boredom .687
My cat is my only friend .680
People fall asleep as soon as I open my mouth to speak .536
I love teaching because students have to pretend to like me or they'll get bad marks .418 .441
Extraction Method: Principal Component Analysis.
Rotation Method: Oblimin with Kaiser Normalization.
a. Rotation converged in 6 iterations.

Looking at the pattern matrix (and using loadings greater than .4 as recommended by
Stevens) we see the following pattern:
Factor 1:
• Q 22. I quiver with excitement when thinking about designing my next experiment
• Q 8. I like control conditions
• Q 17. I enjoy sitting in the park contemplating whether to use participant observation in
my next experiment
• Q 21. Thinking about Bonferroni corrections gives me a tingly feeling in my groin
• Q 13. Designing experiments is fun
• Q 9. I calculate 3 ANOVAs in my head before getting out of bed every morning
• Q 3. I memorize probability values for the F-distribution
• Q 1. I once woke up in the middle of a vegetable patch hugging a turnip that I'd
mistakenly dug up thinking it was Roy's largest root

## Dr. Andy Field Page 7 9/9/2003

Discovering Statistics Using SPSS: Chapter 15

• Q 24. I tried to build myself a time machine so that I could go back to the 1930s and
follow Fisher around on my hands and knees licking the floor on which he'd just trodden
• Q 4. I worship at the shrine of Pearson
• Q 16. Thinking about whether to use repeated or independent measures thrills me
• Q 7. Helping others to understand Sums of Squares is a great feeling
• Q 15. I soil my pants with excitement at the mere mention of Factor Analysis
• Q 11. I like it when people tell me I've helped them to understand factor rotation
• Q 10. I could spend all day explaining statistics to people
• Q 14. I'd rather think about appropriate dependent variables than go to the pub
Factor 2:
• Q 19. I like to help students
• Q 2. If I had a big gun I'd shoot all the students I have to teach (note negative weight)
• Q 6. Teaching others makes me want to swallow a large bottle of bleach because the
pain of my burning oesophagus would be light relief in comparison (note negative
weight)
• Q 18. Standing in front of 300 people in no way makes me lose control of my bowels
(note negative weight)
• Q 26. I spend lots of time helping students
• Q 25. I love teaching
Factor 3:
• Q 5. I still live with my mother and have little personal hygiene
• Q 23. I often spend my spare time talking to the pigeons ... and even they die of
boredom
• Q 28. My cat is my only friend
• Q 12. People fall asleep as soon as I open my mouth to speak
• Q 27. I love teaching because students have to pretend to like me or they'll get bad
marks
No Factor:
• Q 20. Passing on knowledge is the greatest gift you can bestow an individual

This factor is a lot clearer cut: Factor 1 relates to a love of methods and statistics, Factor 2 to
a love of teaching, and factor 3 to an absence of normal social skills. This doesn’t support the
original four factor model suggested because the data indicate that love of methods and
statistics can’t be separated (if you love one you love the other).